Ion-selective Electrode Measurements Research Articles

Interaction of cesium with humic materials: A comparative study of radioactivity and ISE measurements

The interaction of Cs+ with the different fractions of humic materials (humic acid, fulvic acid, and humin) is studied radiometrically and with an ion-selective electrode (ISE). The results indicate that Cs+ can partially complex humic acid, but not fulvic acid. The uptake of Cs+ by humin is small, and the amount of Cs+ required for saturation is 3.72 (mg metal)/(g humin). The ISE is a suitable tool for measuring the free metal ions remaining after the formation of the metal humate complex.

Evaluation of Mitochondrial Function in Isolated Rat Hepatocytes and Mitochondria during Oxidative Stress

The majority of toxic agents act either fully or partially via oxidative stress, the liver, specifically the mitochondria in hepatocytes, being the main target. Maintenance of mitochondrial function is essential for the survival and normal performance of hepatocytes, which have a high energy requirement. Therefore, greater understanding of the role of mitochondria in hepatocytes is of fundamental importance. Mitochondrial function can be analysed in several basic models: hepatocytes cultured in vitro; mitochondria in permeabilised hepatocytes; and isolated mitochondria. The aim of our study was to use all of these approaches to evaluate changes in mitochondria exposed in vitro to a potent non-specific peroxidating agent, tert-butylhydroperoxide (tBHP), which is known to induce oxidative stress. A decrease in the mitochondrial membrane potential (MMP) was observed in cultured hepatocytes treated with tBHP, as illustrated by a significant reduction in Rhodamine 123 accumulation and by a decrease in the fluorescence of the JC-1 molecular probe. Respiratory Complex I in the mitochondria of permeabilised hepatocytes showed high sensitivity to tBHP, as documented by high-resolution respirometry. This could be caused by the oxidation of NADH and NADPH by tBHP, followed by the disruption of mitochondrial calcium homeostasis, leading to the collapse of the MMP. A substantial decrease in the MMP, as determined by tetraphenylphosphonium ion-selective electrode measurements, also confirmed the dramatic impact of tBHP-induced oxidative stress on mitochondria. Swelling was observed in isolated mitochondria exposed to tBHP, which could be prevented by cyclosporin A, which is evidence for the role of mitochondrial permeability transition. Our results demonstrate that all of the above-mentioned models can be used for toxicity assessment, and the data obtained are complementary.

Structure of SDS Micelles with Propylene Carbonate as Cosolvent: a PGSE−NMR and SAXS Study

The effect of propylene carbonate on SDS micelles was investigated by means of pulsed gradient spin-echo (PGSE) NMR, small-angle X-ray scattering (SAXS), conductivity and ion-selective electrode (ISE) measurements. The knowledge of the cosolvent partition between continuous phase and micelles (obtained by means of PGSE-NMR) allowed the identification of relevant dilution paths. Along these paths the system is composed of identical micelles that become more and more diluted. The extrapolation of measured self-diffusion coefficient to infinite dilution (where direct and hydrodynamic interactions are negligible) permits the determination of hydrodynamic size of the micelles. Moreover, the micelle ionization degree (measured by means of ISE) combined with PGSE-NMR and conductivity data furnishes an estimate of the aggregation number without any assumptions on micellar shape. On the other hand, troublesome hydrodynamic interactions are irrelevant to SAXS, and scattering data collected at fixed composition can be analyzed according to a reasonable model by exploiting the insight on the propylene carbonate partition gained through PGSE-NMR. By means of these approaches, we have found that propylene carbonate acts mainly as cosurfactant for the SDS micelles, decreasing their size and aggregation number by increasing the mean headgroup area of SDS.

Transport of chloride and carboxyfluorescein through phospholipid vesicle membranes by heptapeptide amphiphiles.

Seven synthetic anion transporters (SAT) of the general form R(2)N-COCH(2)OCH(2)CO-(Gly)(3)-Pro-(Gly)(3)-OR' were prepared. Three pairs of compounds each contained twin n-hexyl, n-decyl, and n-octadecyl (R) groups at the N-terminus and one contained twin n-tetradecyl groups. Three of the compounds were C-terminated by benzyl and three by heptyl (R') residues. The ability of these compounds to mediate ion release from phospholipid vesicles was assessed. Chloride release was measured by ion selective electrode measurements and by chloride quenching of the fluorescent dye lucigenin. Transport of the anion carboxyfluorescein (CF) was measured by fluorescence dequenching. Differences in both the C- (R') and N-terminal (R) residues within the ionophores affected anion transport. The chloride release data acquired by ion selective electrode and fluorescence methods were similar but not identical. A possible carrier mechanism for Cl(-) transport was discredited. Both Cl(-) and CF anions were released from vesicles by these compounds. The results of CF and Cl(-) transport showed good consistency when the ionophore's N-terminal chains were either decyl or octadecyl but not when they were hexyl. The transport of CF and Cl(-) appears to be fundamentally different when R is C(6) compared to C(10) or C(18). Differences between the behavior of SATs with Cl(-) and CF were also reflected in negative ion mass spectrometric studies.

Comparison of surface and bulk doping levels in chemical polypyrroles of low, medium and high conductivity

AbstractChemical polypyrroles (PPys) of low (σ < 75 S/cm), medium (75 < σ < 200 S/cm) and high electrical conductivity (σ > 200 S/cm) having chloride dopants have been investigated by XPS, chloride ion‐selective electrode (ISE) measurements and high‐resolution termogravimetric analysis (TGA). The average surface doping level in these PPys was 0.28 ± 0.03 as determined from deconvoluted XPS N 1s, Cl 2p and C 1s spectra by using the well‐established N+/N and Cl−/N atomic ratios as well as a new ratio denoted as Cα*/CαTotal. This new ratio provides an estimation of the relative amount of α‐C atoms in charged pyrrole units per total α‐C atoms. The average bulk doping level in these materials was 0.31 ± 0.02 from direct chloride ISE measurements. High‐resolution TGA was employed for the first time in the determination of the amount of hydrogen chloride evolved from the PPy samples during degradation at high temperatures. The resulting average bulk doping level by TGA was 0.30 ± 0.04 for these PPys, in very good agreement with the ISE results. Since surface and bulk doping levels are almost identical for the PPys of low, medium and high conductivity, the differences in conductivity between samples have been attributed to differences in conjugation length among them. For PPy of high conductivity (σ = 288 S/cm), a conjugation length 2.6 times higher than that of PPy of low conductivity (σ = 29 S/cm) has been calculated. Copyright © 2006 John Wiley & Sons, Ltd.

Interactions of bovine lactoferricin with acidic phospholipid bilayers and its antimicrobial activity as studied by solid-state NMR

Bovine lactoferricin (LfcinB) is an antimicrobial peptide released by pepsin cleavage of lactoferrin. In this work, the interaction between LfcinB and acidic phospholipid bilayers with the weight percentage of 65% dimyristoylphosphatidylglycerol (DMPG), 10% cardiolipin (CL) and 25% dimyristoylphosphatidylcholine (DMPC) was investigated as a mimic of cell membrane of Staphylococcus aureus by means of quartz crystal microbalance (QCM) and solid-state 31P and 1H NMR spectroscopy. Moreover, we elucidated a molecular mechanism of the antimicrobial activity of LfcinB by means of potassium ion selective electrode (ISE). It turned out that affinity of LfcinB for acidic phospholipid bilayers was higher than that for neutral phospholipid bilayers. It was also revealed that the association constant of LfcinB was larger than that of lactoferrin as a result of QCM measurements. 31P DD-static NMR spectra indicated that LfcinB interacted with acidic phospholipid bilayers and bilayer defects were observed in the bilayer systems because isotropic peaks were clearly appeared. Gel-to-liquid crystalline phase transition temperatures (Tc) in the mixed bilayer systems were determined by measuring the temperature variation of relative intensities of acyl chains in 1H MAS NMR spectra. Tc values of the acidic phospholipid and LfcinB-acidic phospholipid bilayer systems were 21.5 °C and 24.0 °C, respectively. To characterize the bilayer defects, potassium ion permeation across the membrane was observed by ISE measurements. The experimental results suggest that LfcinB caused pores in the acidic phospholipid bilayers. Because these pores lead the permeability across the membrane, the molecular mechanism of the antimicrobial activity could be attributed to the pore formation in the bacterial membrane induced by LfcinB.

Regarding the Effect that Different Hydrocarbon/Fluorocarbon Surfactant Mixtures Have on Their Complexation with HSA

The complexations between human serum albumin (HSA) and the sodium perfluorooctanoate/sodium octanoate and sodium perfluorooctanoate/sodium dodecanoate systems have been studied by a combination of electrical conductivity, ion-selective electrode, electrophoresis, and spectroscopy measurements. The binary mixtures of the surfactants deviated slightly from ideality. Binding plots revealed the existence of two specific binding sites, the first site being more accessible than the second. Positive cooperative binding has been found, thus revealing the importance of the hydrophobic interactions in both kinds of surfactants. The Gibbs energies of binding per mole of surfactant (DeltaG(nu)) were calculated from the Wyman binding potential where, on the basis of the elevated number of binding sites, a statistical contribution has been included. Initially these energies are large and negative, gradually decreasing as saturation is approached. Changes in the slope of Gibbs energies have been identified with the saturation of the first binding set. These facts denote that the surfactants under study have different favorite adsorption sites along the protein and that the adsorption process of perfluorooctanoate is more closely followed by dodecanoate than by octanoate. Finally, electrophoresis and spectroscopy measurements suggest induced conformational changes on HSA depending on the surfactant mixture as well as the mixed ratio.

Evaluation of Cu–Ethylenediamine Metal Ion Buffers as Calibrants for Ion‐Selective Electrode Measurement of Copper in Fresh Water Systems

AbstractAn investigation was made into the accuracy of cupric ion selective electrode (ISE) measurement of Cu in solutions approximating acidic freshwaters with Cu‐ethylenediamine buffers used as the calibrants. This method overestimates the free Cu compared with calibration using Cu(NO3)2 standards, the standard addition method, and speciation modelling calculations. Statistical tests showed a small, but significant, difference between the intercepts of the linear Nernstian regressions of the calibration plots of Cu‐en buffer standardisation and direct calibration with Cu(NO3)2 standards in matrix that matches the samples. The difference in the intercepts, which corresponds with Eo values of the electrode, is not well understood, but is possibly caused by potentially interfering cations such as Fe2+. The results of this study showed that down to 10−8 M Cu2+, where a linear Nernstian response is possible, the Cu ISE is probably better calibrated using Cu standards prepared in the same matrix as the sample solutions to avoid potential matrix effects.

Evaluation of the Konelab 20XT clinical chemistry analyzer

The Konelab 20XT (Thermo Electron Oy, Finland) is a clinical chemistry analyzer for colorimetric, immunoturbidimetric and ion-selective electrode measurements. The aim of our work was to evaluate the analytical performances of the Konelab 20XT according to the European Clinical Chemistry Laboratory Standards Guidelines. A total of 30 analytes including substrates, enzymes, electrolytes and specific proteins were tested. Investigation results showed low imprecision (within-run coefficient of variation was below 3.5% and between-day coefficient of variation was less than 2.5% for most analytes at all three levels studied) and acceptable accuracy of the analyzer. No significant sample- or reagent-related carry-over was found. It was demonstrated that the analytical system operates within the claimed linearity ranges. The results compared well with those obtained by instruments routinely used in our laboratory (Olympus AU2700, Behring Nephelometer II). In general, the data on interference by hemoglobin, hyperbilirubinemia and turbidity are in accordance with known facts. However, slight hemolysis was found to interfere with the alkaline phosphatase (ALP) assay and mild lipemia affected the glucose assay. The Konelab 20XT is an easy-to-use analyzer that is suitable for routine and emergency analyses in small laboratories.

A Comparison of Copper Speciation Measurements with the Toxic Responses of Three Sensitive Freshwater Organisms

Environmental Context.A rapid Chelex resin method is shown to be a valuable speciation screening tool for use in a tiered risk assessment of copper toxicity in fresh waters. It is a more conservative measure than toxicity testing with sensitive biota, but a better indicator of toxicity than a dissolved copper measurement. Abstract.Twelve natural fresh waters with similar pH and hardness, but varying dissolved organic carbon (DOC) and copper concentrations, were assessed for (a) toxicity to an alga (Chlorella sp. 12), a cladoceran (Ceriodaphnia cf. dubia) and a bacterium (Erwinnia sp.), and (b) copper speciation using a rapid Chelex extraction method, diffusive gradients in thin films (DGT) and anodic stripping voltammetry (ASV). In synthetic fresh water with no added DOC, at pH 7.0 and low hardness, the toxic responses (EC/IC50) of all three organisms to copper were similar. However, in the toxicity of copper added to natural water samples exhibited a negative linear relationship to DOC (r2 = 0.82–0.83), with respective slopes for algae, cladocerans and bacteria decreasing in the ratio 7.4 : 3.5 : 1. The marked difference in responses in the presence of natural dissolved organic matter indicated that not all of the organisms conformed to the free ion activity model (FIAM). This was confirmed by copper ion selective electrode measurement of copper ion activity. Copper toxicity to algae in the presence of DOC was overestimated by free ion activity possibly due to surface binding of DOC. Copper toxicity to the bacteria was greater than predicted and was shown to be a result of bioavailability of some copper complexes formed with organic matter. Cladocerans appear to more closely follow FIAM predictions. These findings have important implications for attempts to extend predictive models of metal toxicity beyond fish to more sensitive freshwater species. The measured labile copper concentrations of copper-spiked natural waters varied from 0 to 70% of total copper concentrations. There was no clear relationship between the three measurement techniques. Good correlations were obtained between both algal and bacterial growth inhibitions measured on copper-spiked natural waters and the corresponding Chelex-labile copper concentrations. A single natural water sample was manipulated to different pH and hardness values, spiked with copper, and tested using the above three organisms with the Chelex method. Toxicity test results generally agreed with studies performed in synthetic fresh waters, showing that the relationships between toxicity, pH and hardness were organism-specific. Chelex-labile copper was always over-predictive of toxicity but significantly better (P ≤ 0.05) than dissolved copper concentrations, as it only detects the fraction of total copper that is reactive over biologically-relevant timescales. Colloidally-bound copper and copper associated with strong ligands are not detected. The Chelex method is therefore useful as a measure where speciation is accepted in water quality regulations.

A comparative study of ordinary and mineralised Portland cement clinker from two different production units: Part I: Composition and hydration of the clinkers

Portland cement clinkers from two production units were investigated; Plant 1: ordinary clinker (P1) and clinker mineralised with CaF 2+CaSO 4 (P1m); Plant 2: ordinary clinker (P2) and two clinkers mineralised with CaF 2+CaSO 4 (P2m, low SO 3 and P2m′, high SO 3). The chemical composition of the clinkers was determined by X-ray fluorescence, ICP analysis, titration (free lime) and ion selective electrode measurements (F). Observed clinker parameters (LSF, SR, AR, R, wt.% MgO, F, SO 3, free lime): P1 (0.96, 2.72, 1.27, 1.04, 0.78, 0.06, 0.64, 0.71); P1m (1.03, 2.21, 1.58, 2.18, 0.87, 0.23, 1.95, 0.69); P2 (1.00, 2.66, 1.72, 0.75, 4.06, 0.20, 1.38, 1.51); P2m (1.01, 2.91, 1.96, 0.90, 3.21, 0.39, 1.72, 2.06); P2m′ (0.97, 2.70, 1.84, 1.15, 3.86, 0.42, 2.48, 0.89). The qualitative and quantitative phase compositions were characterised using X-ray powder diffraction, backscattered electron imaging, X-ray microanalysis and elemental mapping, plus optical reflection microscopy. Phases observed in all clinkers were: alite, β-belite, cubic aluminate, ferrite and free lime. Additional phases observed were: aphthitalite (P1, P2, P2m, P2m′), calcium langbeinite (P1m) and periclase (P2, P2m, P2m′). The clinker composition and texture differ more between the two plants, than between ordinary and mineralised clinker from the same production unit. Laboratory cements were prepared by mixing ground clinker with CaSO 4·2H 2O. The cements were hydrated in an isothermal calorimeter at 20 °C (water/cement weight ratio=0.5) during 33 h. After 12 h, the laboratory cement based on P1m reached a higher level of reaction than the one based on P1. The P2m and P2m′ laboratory cements had a slower reaction than the P2 cement.

The influence of sewage sludge properties on sludge-borne metal availability

With the advent of more stringent controls on wastewater treatment, sewage sludge production in Europe and many parts of the world is increasing. With this increase comes the problem of sludge disposal, and recycling to land arguably offers an economically and environmentally sustainable option. However, a major limitation of sewage sludge reuse is the potential release of heavy metals from the sludge and heavy metal accumulation to toxic levels in topsoils. The properties of the sludge play a crucial role in determining the initial release and subsequent availability of heavy metals in amended soils. Bioavailable forms of heavy metals in recently amended soils are most likely to be those that are bioavailable in the sewage sludge. In this paper, published research on the importance of sewage sludge characteristics on metal release and bioavailability will be reviewed and contrasted with original research. A selection of sludges from around Australia has been collected for this purpose. Through the use of incubation studies, isotope dilution techniques, ion-selective electrode measurements and 13C-NMR spectroscopy, the importance of a range of sludge properties on heavy metal behaviour in sludges and sludge-amended soils is addressed.

Characterization of copper complexation with natural dissolved organic matter (DOM)—link to acidic moieties of DOM and competition by Ca and Mg

We investigated Cu complexation by three dissolved organic matters (DOMs) collected by reverse osmosis (RO). Alkalimetric titration, pH-stat Cu and Ca titrations, pH edges of Cu–DOM complexation, and Ca/Mg–Cu exchange experiments were investigated at I=10 −2 M for DOM samples of 10 mg C/L. The proton and Cu binding characteristics indicated similarity for all three DOMs. All Cu titrations employed ion selective electrode measurement and indicated the presence of relatively small amounts of strong Cu-binding sites. Four distinct classes of Cu binding sites are required for FITEQL 4.0 to provide good fits to the entire curves. The estimated total Cu binding site density is 4.55 mmol/g C, much less than the total acidity but very close to the phenolic site content. Cu–DOM complexation increases approximately 10-fold per pH unit, even at relatively high pH (>8). We suggest that sites characterized as phenolic based on alkalimetric titration, not carboxyl sites, account for the majority of Cu complexation under natural water conditions, and Cu–DOM complexation is principally through the replacement of H + by Cu 2+ at the phenolic binding sites. The Cu–H exchange ratio is 1:1 for the first three sites and about 1:2 for the 4th site. This 4-site model describes well the pH dependency of Cu–DOM complexation and provides good estimates of free Cu concentrations throughout wide total copper (Cu T) and pH ranges. Comparison between Ca–DOM and Cu–DOM complexation demonstrated that (i) Ca–DOM complexation increases much less than an order of magnitude per pH unit and decreases at higher Ca concentration, different from that of Cu–DOM complexation; and (ii) Cu–DOM complexation is highly non-linear, in contrast to the much reduced extent of non-linearity of Ca–DOM complexation. Ca/Mg–Cu exchange experiments showed small competition effect, less than expected by a simple competition model, and the competition tended to reduce with increasing Ca or Mg concentrations. The extent of the competition by Mg and Ca are essentially comparable. Put all together, it suggests that Ca and Mg are preferably bound by carboxyl sites, especially at relatively high concentrations, resulting in a weakened apparent competition effect.

Comparison of Enzymatic and ISE Methods for Ammonia Measurements in Untreated Red and White Juices and Wines

The use of a standard, commercial enzymatic assay for determination of ammonia in juices and wines, with a focus on red juices and wines, has been investigated. Ammonia standards were prepared in model wine solutions, white juice and wine, and red juice and wine that ranged from 0 to 300 mg/L of added ammonia. Ammonia concentration in all standards was then determined by enzymatic assay and by ion-selective electrode for comparison. Repeatability and linearity of the enzymatic assay were excellent for both red juice and wine with only a simple filtration step as a pretreatment. Replicate measurements had a coefficient of variation of less than 2%, and the correlation coefficients (R) for both juice and wine were greater than 0.99. These results were similar to those found for the ion-selective electrode measurements. Acknowledgment: The authors wish to acknowledge the American Vineyard Foundation for funding this work through the grant, Evaluation of the Influence of Common Viticultural Practices on the Chemical and Sensory Characteristics of Wines.

Graphical methods of determination of stability constants of metal complexes using spectrophotometric and ion-selective electrode measurement data: cobalt(II)-PAR complex

A new heterogeneous Co I I -ion selective electrode has been developed and its characteristics have been studied. Using the electrode the stability constant of the complexes formed by Co I I -ion with 4-(2-pyridylazo)resorcinol (PAR) has been determined (log β 2 = 19.75) involving a graphical method. Stability constant of 1: 2 complexes, determined by the ion-selective electrode (log β 2 = 19.75) agrees well with the value (log β 2 = 19.6) determined by spectrophotometric method.

Standardization of the Ammonia Electrode Method for Evaluating Seafood Quality by Correlation to Sensory Analysis

ABSTRACT: Ammonia ion‐selective electrode (ISE) measurements, reported as apparent ammonia, were correlated to expert sensory assessments of 6 different fish species stored on ice and at room temperature. Total volatile base nitrogen (TVB‐N), trimethylamine nitrogen (TMA‐N), and apparent ammonia showed the same development trend during storage. ISE measurements and TVB concentrations had a correlation of r2= 0.92. Sensory assessment, using a 1‐ to 100‐mm line scale with values > 50 considered unacceptable, resulted in an r2between sensory scores and ISE measurements of 0.78. Regression analysis predicted 19.6 mg/100 g of apparent ammonia in fish tissue at the sensory limit of 50, regardless of storage conditions. ISE measurements could be used in predicting borderline quality and decomposition.

Cyanide ion-selective electrode measurements in the presence of copper

The response of a cyanide ion-selective electrode (ISE) in the presence of copper has been investigated. The interferences of metal ions, like copper, on the response of the cyanide ISE prevents its use for routine direct analysis of metal cyanide containing waste water. In this work an improved understanding of the chemistry at the electrode, and the effects of different cyanide and copper concentrations and matrix pH values on the electrode response, has been developed. Based on this insight, a semi-empirical approach is described to allow calibration of the ISE enabling the estimation of total cyanide. While not as robust as, for example, total cyanide distillation, it provides an inexpensive, rapid and convenient approach for more frequent and widespread screening of industrial waste waters.

Analytical methods and quality assurance

The development of strategies to improve the efficiency of nutrient uptake by crops and to minimise leaching to surface and drainage water needs to be considered at various scales: catchment, field, soil and crop type. Therefore, methodological approaches for routine sample preparation and analysis are crucial to agricultural and environmental policies aimed at reducing nutrient losses. While the official way of analysing water is time consuming and requires a laboratory with standard equipment, the ion‐selective electrodes (ISE) technique presents a method with the advantages of precision, speed and portability. Analysis of water by ISE has become an accepted practice for the detection of certain elements and this technique could replace standard testing currently being used. Accuracy and reproducibility of NO3 ‐, Cl‐, Ca+2, K+ and Na+ ion‐selective electrodes were achieved by comparison with standard methods. The apparatus used consists of a sensor unit, which contains multielement ISE's, and a mesurement unit for operational and data storage functions. Samples of water, obtained from runoff and drainage of agricultural catchments, ground water, were selected from different locations to represent a wide range of ion concentrations. Using the ISE measurements values varied from 0.42 to 83.5 (mg L‐1) for NO3 ‐, 3.89 to 241 for Cl‐, 1.3 to 76.4 for Ca2+, 0.12 to 131 for K+ and 0.63 to 59 for Na+. The investigation confirms the good correlation between standard and ISE methods for almost all ions with correlation coefficients of 0.97 for NO3 ‐, 0.99 for Cl‐, 0.96 for Ca2+, 0.98 for K+. Only Na+ coefficient is below 0.9. Variations in the slope and intercept of the regression equations were slight or small for NO3 ‐ and Ca2+, and larger for Cl‐, K+ and Na+.

Amylose‐iodine complex formation without KI: Evidence for absence of iodide ions within the complex

Amylose-iodine (AI) complex has been synthesized in aqueous solution without added KI. Complex formation (with solid iodine in amylose solution) is maximized at approximately 35°C and decreases beyond that temperature. Ion-selective electrode (ISE) measurements of an aqueous solution of iodine and AI complex indicate that there is no change in the I− ion concentration when the complex forms. This suggests that I− ions (including I, I, and others) cannot be involved in forming the AI complex. The present work also reports a new and simple method for providing both the iodine-binding capacity (IBC) of amylose and the dissociation mechanism for the AI complex. © 1999 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 37: 2711–2717, 1999

Amylose-iodine complex formation without KI: Evidence for absence of iodide ions within the complex

Amylose-iodine (AI) complex has been synthesized in aqueous solution without added KI. Complex formation (with solid iodine in amylose solution) is maximized at approximately 35°C and decreases beyond that temperature. Ion-selective electrode (ISE) measurements of an aqueous solution of iodine and AI complex indicate that there is no change in the I− ion concentration when the complex forms. This suggests that I− ions (including I, I, and others) cannot be involved in forming the AI complex. The present work also reports a new and simple method for providing both the iodine-binding capacity (IBC) of amylose and the dissociation mechanism for the AI complex. © 1999 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 37: 2711–2717, 1999